Abiray/K2-Horizon-MoVA-36B-A4B-GGUF
41.1k
K2-Horizon-MoVA-36B-A4B-GGUF
This repository contains GGUF quantizations of IFM/K2-Horizon-MoVA-36B-A4B, a sparse Mixture-of-Experts model utilizing Mixture-of-Values Attention (MoVA). It features 36B total parameters with only ~4B activated per token, enabling near-small-model inference speeds with 36B-scale intelligence.
[!IMPORTANT] Architecture Support Required: As of early 2026,k2-horizonis a newly introduced architecture not yet merged into upstream mainlinellama.cpp. To run or quantize these GGUFs, you must build from the official MBZUAI-IFM fork (model/K2Horizonbranch). Running on vanillallama.cppwill result inunknown model architecture: 'k2-horizon'.
Quantization Overview & Hardware Recommendations
Installation & Setup
1. Build llama.cpp with K2-Horizon Support
# Clone the official architecture fork
git clone -b model/K2Horizon [https://github.com/MBZUAI-IFM/llama.cpp.git](https://github.com/MBZUAI-IFM/llama.cpp.git)
cd llama.cpp
# Build with hardware acceleration (enable CUDA if running on NVIDIA GPUs)
cmake -B build -DGGML_NATIVE=ON -DCMAKE_BUILD_TYPE=Release # Add -DGGML_CUDA=ON for NVIDIA GPU
cmake --build build -j$(nproc) --target llama-cli llama-serverQuickstart Guide
Running via llama-cli
./llama.cpp/build/bin/llama-cli \
-m ./K2-Horizon-MoVA-36B-A4B-Q4_K_M.gguf \
-ngl 99 \
-c 8192 \
--temp 1.0 \
--top-p 0.95 \
-p "<|ifm|im_start|>user\nWrite an efficient Python script for asynchronous data fetching.<|ifm|im_end|>\n<|ifm|im_start|>assistant\n"Prompt Format & Reasoning Traces
K2-Horizon uses ChatML-style delimiters with native thinking tags:
<|ifm|im_start|>system
You are a helpful assistant.<|ifm|im_end|>
<|ifm|im_start|>user
Your query goes here.<|ifm|im_end|>
<|ifm|im_start|>assistant
<ifm|think>
[Model generates intermediate reasoning steps here]
</ifm|think>
[Final output generated here]<|ifm|im_end|>